AMBER Archive (2006)

Subject: Re: AMBER: vlimit exceeded for step 0, vmax = infinity

From: Carlos Simmerling (carlos_at_csb.sunysb.edu)
Date: Wed Nov 08 2006 - 11:46:27 CST

I think I've seen this recently and I have some idea what
might be happening but I need to run more tests and check
the code. Can you send me your output files for the
run that crashed and the run right before it? Send them
directly so we don't clutter people's mail.
thanks
carlos

===================================================================
Carlos L. Simmerling, Ph.D.
Associate Professor Phone: (631) 632-1336
Center for Structural Biology Fax: (631) 632-1555
CMM Bldg, Room G80
Stony Brook University E-mail: carlos.simmerling_at_stonybrook.edu
Stony Brook, NY 11794-5115
===================================================================

Rachel wrote:

> Hi, David,
>
> Thanks for your advice, i tried to change the tempi as well as the ntt
> value as you suggested, however, both no luck, i still got the same
> error message, and as I wrote in another email, i noticed that the
> bond energy from the last step of minimization decreased dramatically
> when I started to run the equilibration.
>
> Here is the output of the last step of the minimization:
> ---------------------------------------------------------------------------------------------------------------
>
> NSTEP ENERGY RMS GMAX NAME NUMBER
> 5000 -2.0193E+05 5.2652E-01 7.9133E+01 FE 12110
>
> BOND = 13557.7748 ANGLE = 2100.0258 DIHED =
> 6683.4797
> VDWAALS = 24316.5978 EEL = -284589.3077 HBOND =
> 0.0000
> 1-4 VDW = 2495.1800 1-4 EEL = 33508.5993 RESTRAINT =
> 0.0000
> ------------------------------------------------------------------------------------------------------------------
>
> and here is the output of the equilibration of the system with
> restraints on the solute:
>
> -------------------------------------------------------------------------------------------------------------------
> | Local SIZE OF NONBOND LIST = 17919959
> | TOTAL SIZE OF NONBOND LIST = 17919959
>
> NSTEP = 0 TIME(PS) = 0.000 TEMP(K) = 0.00 PRESS
> = 0.0
> Etot = -215088.9939 EKtot = 0.0000 EPtot =
> -215088.9939
> BOND = 396.4313 ANGLE = 2100.0258 DIHED =
> 6683.4797
> 1-4 NB = 2495.1800 1-4 EEL = 33508.5993 VDWAALS =
> 24316.5978
> EELEC = -284589.3078 EHBOND = 0.0000 RESTRAINT
> = 0.0000
> Ewald error estimate: 0.1110E-03
>
> vlimit exceeded for step 0 ; vmax = Infinity
>
> Coordinate resetting (SHAKE) was not accomplished
> within 3000 iterations
>
> Note: This is usually a symptom of some deeper
> problem with the energetics of the system.
>
> ---------------------------------------------------------------------------------------------------------------------
>
> As can be seen, the bond energy changed from 13557.7748 to 396.4313
> while all other energies remained the same, is this happening because
> I changed the ntc, ntf value from the minimization to the md?
>
> Can anyone suggest how to solve this problem please?
>
> Thanks in advance!
>
> Rachel
>
>
>
>
> On 11/8/06, *David A. Case* <case_at_scripps.edu
> <mailto:case_at_scripps.edu>> wrote:
>
> On Wed, Nov 08, 2006, Rachel wrote:
>
> > vlimit exceeded for step 0 ; vmax = Infinity
> >
> > tempi = 0.0,
> > temp0 = 300.0,
> > ntt = 3,
> > gamma_ln = 1.0,
>
> As a wild guess, try a short (10 step) simulation with tempi set
> to some small
> but non-zero value. It is possible that Langevin dynamics has
> some bad
> behavior when all the velocities are zero. Alternatively, try a short
> simulation with ntt=1.
>
> This is shooting in the dark, but it is simple enough to be worth
> a try.
>
> ...good luck....dac
>
> -----------------------------------------------------------------------
>
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu <mailto:amber_at_scripps.edu>
> To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
> <mailto:majordomo_at_scripps.edu>
>
>

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu